VIROLOGY

174,9-l

7 (1990)

Administration of Noninternal Image Monoclonal Anti-idiotypic Antibodies Induces Idiotype-Restricted Responses Specific for Human lmmunodeficiency Virus Envelope Glycoprotein Epitopes EN-MIN ZHOU,’ KENTON L. LOHMAN, AND RONALD C. KENNEDY’ Department

of Virology and immunology,

Southwest

Foundation

for Biomedical Research, San Antonio,

Received June 1, 1989; accepted September

Texas 78284

19, 1989

A mouse monoclonal anti-idiotypic antibody (anti-Id), designated MC1, was generated against chimpanzee antibodies specific for a synthetic peptide corresponding to a native epitope associated with gp41 of human immunodeficiency virus (HIV). This anti-Id recognized a shared idiotope/idiotype (Id) on a second chimpanzee anti-gp41 peptide preparation but failed to detect this Id on rabbit and mouse anti-gp41 peptide antibodies induced by immunization with the gp41 synthetic peptide. The chimpanzee Id-MC1 reaction was not inhibited by either synthetic peptide or recombinant gp160 suggesting that MC1 exhibits noninternal image, AbQa-like characteristics. Immunization of syngeneic Balb/c mice with MC1 induced an antigen-positive (Ag+) response capable of binding the synthetic peptide, recombinant gpl60, and gp41, whereas MC1 -immunized rabbits did not produce any detectable anti-peptide and/or anti-HIV envelope glycoprotein antibody response. The MCl-induced anti-Id response (Ab-3) in both mice and rabbits expressed a similar Id with the Ab-1, which is not normally expressed in the anti-gp41 peptide antibody response induced by the nominal antigen in Balb/c mice and in rabbits. Together, these studies indicate that a mouse monoclonal anti-Id of the Ab-Pa class can induce an anti-HIV response specific for a gp41 epitope defined by a synthetic peptide, which does 0 1990Academic Press, Inc. not Cross SpeCieS barriers.

recognize Ab-1 preparations from other species. The Ab-2y subclass recognizes an antigen combining site Id but is not an internal image. The human immunodeficiency virus type 1 (HIV-l) envelope gene product is synthesized within infected cells. This glycosylated polyprotein with an MW of 160,000 (gpl60) is processed into subunits with MW of 120 kDa (gpl20) and 41 kDa (gp41) (Allan et a/., 1985; DiMarzo et al., 1985). In previous studies, our laboratories have identified potential antigenic determinants using synthetic peptides analogous to regions of the envelope glycoprotein precursor associated with gp41 (Pauletti eta/., 1985; Kennedy eta/., 1986d). Rabbit antisera against a synthetic peptide analogous to amino acid sequences 735 to 752 of the gp160 from the human T-cell leukemia virus type III B (HTLV-IIIB) isolate of HIV-1 have been shown to neutralize the infectivity of the HIV-l isolate HTLV-IIIB and of NY-5, a divergent isolate, in vitro (Chanh et a/., 1986). In this report, we describe the generation and characterization of mouse monoclonal anti-Id, designated MC1 , which was produced against chimpanzee antigp41 antibodies induced by immunization with synthetic peptide 735 to 752. This anti-Id has been classified as an Ab-2a since (i) neither the synthetic peptide nor a recombinant gpl60 antigen could inhibit the Idanti-Id reaction, and (ii) the anti-Id failed to recognize the Id expressed on anti-gp41 peptide antibodies in-

INTRODUCTION Perhaps the most significant role of an anti-idiotypic antibody (anti-Id) involves its potential regulation of the immune response to a given antigen. Experimentally, the ability of anti-Id to manipulate the immune responsein viva has been documented in numerous antibody systems .(Uytehaage et a/,, 1986; Zhou et al., 1987b) and gives clear support for Jerne’s idiotype (Id) network theory of immune regulation (Jerne, 1974). Included in this modulation of the immune response are the principles associated with the use of anti-Id as putative vaccines. According to Jerne, two kinds of antiId (Ab-2) can be distinguished by serological means: an Ab-2a and an Ab-2/3 (Jerne et a/., 1982). An alternative classification of anti-Id was proposed by Bona and Kohler (1984) that subgroups the Ab-2cu and Ab-2P classes into an Ab-2a, Ab-2/3, and Ab-27. Ab-2a recognizes a nonantigen combining site Id or nonantigen inhibitable Id and usually recognizes a private Id. Ab-2& also referred to as internal image anti-Id, has the capacity to mimic the antigen used to generate the Ab-1 and can substitute for the nominal antigen in inducing an anti-antigen response. In addition, the Ab-2/3 will ’ Present address: Immunology Department, University of Manitoba, Winnipeg, Manitoba, Canada. ‘To whom correspondence and reprint requests should be addressed. 9

0042-6822/90

$3.00

Copyright@ 1990 by Academic Press. Inc All rights of reproduction in any form reserved.

ZHOU, LOHMAN,

10

duced in other species (rabbits and mice) by immunization with the synthetic peptide. Immunization of Balb/c mice with MCI induces an anti-peptide and anti-HIV response, whereas rabbits injected with MC1 failed to produce any detectable anti-peptide and/or anti-HIV response. This suggests that the anti-gp41 antibody response (Ag+) induced by McAb MC1 is restricted to mice. However, both mice and rabbits produced an anti-Id response to MCI, and these polyclonal antiMC1 antibodies (Ab-3) share a similar Id expressed on the chimpanzee Ab-1 preparation, but not on mouse or rabbit anti-gp41 antibodies induced by immunization with the synthetic peptide (nominal antigen). These results indicate that monoclonal Ab-2a anti-Id can activate Id-specific clones which exhibit shared Id in both syngeneic and xenogeneic immune responses. However, the antigen-specific component of the response recognizing HIV-1 epitopes was induced only in the syngeneic system.

AND KENNEDY

was precipitated with 18% sodium sulfate (w/v) and affinity purified on peptide 735 to 752 containing immunoadsorbent columns. Five molar potassium iodide was employed to elute the anti-peptide antibody fractions which were then pooled and exhaustively dialyzed against borate-buffered saline (BBS). The IgG was purified from the resulting anti-peptide preparation by Sephadex G-200 gel filtration. Mouse monoclonal antibodies (IgM) were produced by fusing the spleen cells of mice immunized with the peptide coupled to KLH (Dalgleish et a/., 1988). All antibody preparations recognized gp41 by Western blot analysis. Preparation

of monoclonal

antibody

(McAb) anti-Id

Adult female Balb/c ByJ mice and New Zealand white rabbits were purchased from Jackson Laboratories, Bar Harbor, ME, and from Hazelton Laboratories, Denver, PA, respectively.

Female Balb/c ByJ mice were immunized with affinity purified chimpanzee IgG antibodies directed against synthetic peptide 735 to 752. Mice were given 20 pg of chimpanzee antibodies intraperitoneally (i.p.) in Freund’s complete adjuvant biweekly. Mice were boosted intravenously with 5 pg of chimpanzee antibodies 3 days prior to fusion. Hybridoma production has been published in detail elsewhere (Chanh et al., 1987). Supernatant fluids were screened in a biotinavidin-amplified enzyme-linked immunosorbent assay (ELISA) which is described in detail below. Those supernatant fluids which were positive for anti-id activity were subcloned by limiting dilution.

Synthetic

Detection

MATERIALS

AND METHODS

Animals

peptides and HIV antigens

The synthesis of peptide 735 to 752 corresponding to gp160 amino acid sequences from the HTLV-IIIB isolate of HIV-1 is described in detail elsewhere (Chanh eta/., 1986; Kennedy et a/., 1986d). A purified recombinant gp160 peptide produced in the baculovirus expression vector system was the gift of Dr. Gale Smith, MicroGeneSys, Inc., West Haven, CT. The control peptides used in this study correspond to simian virus 40 (SV40) large tumor antigen (T-ag) nuclear transport peptide (Lanford et al., 1986) and a synthetic peptide analogous to amino acid sequences from the S region of hepatitis B surface antigen (HBsAg). Antibodies

and their purification

The generation, purification, and characterization of chimpanzee, rabbit, and mouse anti-gp41 antibodies induced by immunization with peptide 735 to 752 have been published previously (Kennedy et a/., 1986a,d, 1987a; Dalgleish et a/., 1988). The immunization consisted of synthetic peptide 735 to 752 coupled to keyhole limpet hemocyanin (KLH) and emulsified in Freund’s adjuvant (Difco Laboratories, Detroit, Ml). Serum from individual chimpanzees and rabbits was collected. The pooled individual anti-peptide antiserum

of McAb anti-Id

The McAb anti-Id, designated MC1 , was subcloned by limiting dilution and characterized serologically with a biotin-avidin-amplified ELISA (Schick et a/., 1987; Zhou eta/., 1987a). Briefly, 200 ng in 50 ~1of chimpanzee anti-gp41 peptide antibodies along with control chimpanzee affinity purified IgG anti-SV40 T-ag peptide antibodies and chimpanzee preimmune IgG were adsorbed to wells of microtiter plates (Immulon II; Dynatech Laboratories, Inc., Alexandria, VA). Nonspecific sites were blocked with 10% normal goat serum (10% NGS). Fifty microliters of supernatant or ascites fluid containing the anti-Id MC1 was added to the chimpanzee antibody-coated microtiter wells. Specific binding was determined using biotinylated goat anti-mouse yglobulin (b-gamgg) and streptavidin-conjugated horseradish peroxidase. The substrate was 2,2’-azino-di-(3ethyl-benzthiazoline sulfonic acid (ABTS) and 0.01% H202. The calorimetric reaction was stopped by adding 10% sodium dodecyl sulfate (SDS). Optical density of each well was read on an automatic ELISA plate reader at 410 nm (OD,,,). The hybrid clone which recognized chimpanzee anti-gp41 antibodies, yet failed to bind control chimpanzee anti-peptide or preimmune chimpanzee tgG, was designated MCI. Ascites fluid

ANTI-Id

INDUCES

IMMUNE

was generated by i.p. injection of 5 x 1O6 hybrid cells into pristane (Aldrich Chemical Co., Milwaukee, WI)primed Balb/c ByJ mice. The purification of mouse McAb from ascites fluid using a combination of salt fractionation and molecular sieve chromatography has been described in detail elsewhere (Kennedy et al., 1983). The isotype of MC 1 was determined to be IgG1 by using murine isotype-specific antibody reagents (Bio-Rad Laboratories). Characterization of the anti-id, MC1 The anti-Id specificity of MC 1 was determined by using the direct binding ELISA with chimpanzee IgG antigp41 peptide along with control and preimmune chimpanzee IgG and various concentrations of purified MC 1 as described above. To further characterize the anti-Id, the following inhibition of binding assays were performed. Inhibition of the Id-anti-Id reaction was used to examine the inhibiting capacity of various chimpanzee, rabbit, and mouse anti-gp41 peptide antibodies. A constant amount of chimpanzee IgG anti-gp41 peptide antibodies (prototype Id antibody designated X-99) (200 ng in 50 ~1)was adsorbed to the solid phase of an ELISA plate overnight at 4”. Nonspecific binding sites were blocked by adding 10% NGS for 30 min at 37” and then the plate was washed in T-PBS. Biotinylated-anti-Id (b-MCI; 30 ng per well) was mixed with different concentrations of various inhibitors and then added to triplicate wells and allowed to incubate for 1 hr at 37”. The remainder of the assay is as described previously (Schick et al., 1987). Maximum binding (no inhibition) was determined when b-MC1 was added without inhibitor. Background OD,,, values (nonspecific binding resulting from OD values obtained in microtiter wells that were not coated with antigen) were subtracted from each specific value. In order to ascertain whether the anti-Id could inhibit the Id from binding to either peptide 735 to 752 or recombinant gpl60, a competitive inhibition assay was performed. Inhibition of the Id (chimpanzee X-99 antigp41 peptide) from binding either peptide 735 to 752 or gpl60 was examined in a manner similar to the inhibition of Id-anti-Id interaction with the following modifications: 200 ng of peptide 735 to 752 was adsorbed to the solid phase overnight at 4”. Biotinylated-Id (100 ng in 50 ~1)was then mixed with various concentrations of MC1 and other inhibitors and then added to triplicate wells in microtiter plates. The addition of the substrate and quantitation of the reaction is as described above. Percentage inhibition of Id-anti-Id or Id-peptide reaction was calculated with the following formula: % inhibition = 100 1 _ (OD,,, with inhibitor - background) (OD,,, without inhibitor - background ’

1

RESPONSE TO HIV

KLH coupling, of antibodies

11

alum precipitation,

and biotinylation

The methods for coupling of antibodies to KLH along with alum precipitation have been described in detail elsewhere (Kennedy and Dreesman, 1984; Schick et a/., 1987). Briefly, antibodies (IgG) in a concentration of 5 mg/ml cooled to 4’were mixed with a 10,000: 1 molar ratio of 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDAC) (Sigma Chemical Co., St. Louis, MO) to IgG. After stirring at 4” for 30 set, KLH at a molar ratio of 1: 50 KLH to IgG was added to the IgG and stirred for 2 hr at 25” and then mixed overnight at 4”. The IgG-KLH conjugate was dialyzed against BBS overnight at 4” and then alum precipitated using aluminum potassium sulfate. Biotinylation of various Id and anti-Id preparations have been described by this laboratory previously (Schick et a/., 1987). ln vivo administration and rabbits

of anti-Id MC1 in Balbk mice

Female Balb/c ByJ mice and female New Zealand white rabbits were given six injections every 14 days of anti-Id MC1 (20 pg for mouse and 200 pg for rabbit) as an alum precipitate or conjugated to KLH and as an alum precipitate. Animals were bled at 14 and 28 days after the final immunization. Sera obtained from mice and rabbits prior to immunization served as negative control. All rabbit antisera were repeatedly adsorbed on a Sepharose 4B (Pharmacia) column containing a pool of normal mouse y-globulin until all detectabte anti-isotype and anti-allotype antibodies were removed. The assays for detecting rabbit anti-Id (antiMCl) to a mouse McAb have been described in detail elsewhere (Schick eta/., 1987). Determination

of anti-peptide

and anti-HIV activity

The anti-peptide 735 to 752 response in mouse or rabbit antisera from animals immunized eitherwith 735 to 752 coupled to KLH or MC1 was determined by using a biotin-avidin-amplified ELISA with b-gamgg and biotinylated goat anti-rabbit y-globulin (b-gargg) as probes (Kennedy et a/., 1986d). Anti-peptide levels in mouse and rabbit antisera were quantitated as an end point titer and represent the last dilution of anti-peptide containing sera that resulted in an OD,,, three times higher when compared with the OD,,, obtained with a 1: 10 dilution of the preimmune sera. The anti-HIV activity in the mouse and rabbit antisera was similarly determined using a baculovirus gpl60 recombinant preparation. Twenty nanograms in 50 ~1 of recombinant gpl60 was adsorbed onto the microtiter plate overnight at 4’. The remaining assay was per-

12

ZHOU, LOHMAN.

formed as described previously (Zhou et al., 1987a). The anti-HIV titers are similarly expressed as end point titer dilution. Background OD,,, values were subtracted from each individual value as described above. Preparation

AND KENNEDY

5 =

0.8

of immunoadsorbents

The methods for the coupling of peptide 735 to 752 and SV,, T-ag transport peptide or normal mouse yglobulin to cyanogen bromide-activated Sepharose 4B and Affi-Gel 501 have been described in detail previously (Zhou et al., 1987a). Western blot analysis Western blot analysis was performed using the method which has been described in detail previously (Barin et a/., 1985). Briefly, the culture supernatant of HTLV-IIIB-infected Molt-3 cells was fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The protein in the gel was passively transferred to nitrocellulose which was then nonspecifically blocked with 3% BSA at 37” for 1 hr. The nitrocellulose strips were washed in T-PBS and incubated with pooled human AIDS (a positive control; 1: 100) and mouse sera, and rabbit sera at a dilution of 1:20 at 37” for 2 hr. Biotinylated goat anti-human, anti-mouse, antirabbit y-globulin, and streptavidin horseradish peroxidase (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MA) were used to detect human, mouse, and rabbit antibody reactivities. Diaminobenzidine was used as the substrate. RESULTS Mouse McAb anti-Id recognizes

Id determinants

The specificity of mouse McAb anti-Id MCI was determined by direct binding assays utilizing the Id along with other chimpanzee IgG preparations. Two hundred nanograms of various chimpanzee IgG were adsorbed onto the microtiter plates and various concentrations of MC1 and an irrelevant mouse IgGl anti-Id (a McAb specific for the HBsAg system) were added. The results are depicted by the binding curves in Fig. 1. At a concentration of 5 pg/ml (final concentration of 250 ng in 50 PI), MC1 bound to chimpanzee X99 and Xl83 anti-gp41 peptide IgG with an OD,,, value of 0.72 and 0.69, respectively. Higher concentrations of MC1 did not increase the OD value as it relates to the specific binding to these Id preparations. There was no binding to IgG from chimpanzee Xl 89 anti-SV40 T-ag peptide or chimpanzee X99 and Xl83 prior to immunization with their respective peptides. Also, no binding was observed between irrelevant mouse IgGl and the chimpanzee IgG preparations (data not shown). These re-

CONCENTRATION

OF ANTI-ID

( FgIml)

FIG. 1. Binding curves of mouse monoclonal anti-Id (MCI) with various chimpanzee antibody preparations. MC1 binding IgG from chimpanzee 99 anti-HIV gp41 peptide (O), chimpanzee 183 anti-HIV gp41 peptide (0); chimpanzee 189 anti-W40 peptide (Cl); chimpanzee 99 preimmune IgG (A); chimpanzee 183 preimmune IgG (V). Each point represents the mean of triplicate values. The vertical bars are representative of the range of OD values.

sults suggest that MC1 recognizes an Id determinant shared by two chimpanzee anti-gp41 peptide preparations since no reactivity was demonstrated with control chimpanzee IgG preparations. Mouse McAb anti-Id recognizes determinants on the Id

noncombining

site

An inhibition assay was employed to examine whether the anti-Id MC1 could inhibit the chimpanzee anti-gp41 Id or recombinant gpl60 from binding to HIV peptide 735 to 752. At various concentrations (ranging from final concentrations of 200 pg to 20 ng), MC1 failed to inhibit chimpanzee Ab-I from binding to either the peptide or recombinant HIV gpl60 antigens (data not shown). We could demonstrate that the chimpanzee Ab-1 bound MC1 at the same time that the Ab-1 was bound to antigen in this inhibition assay by utilizing b-gamgg. Confirming this observation, no inhibition of the Id-anti-Id reaction was observed with either the peptide or the recombinant gpl60 antigens (Table 1). Together, these results suggest that anti-Id MC1 recognizes an antigen noncombining site Id which is shared by another chimpanzee Id and is potentially located within the framework region of the specific antigp4 1 Ab-1 preparations. Mouse McAb anti-Id fails to recognize interspecies Id

an

In order to detect whether the anti-Id could recognize Ab-1 preparations of similar specificity produced in

ANTI-Id

INDUCES

IMMUNE

TABLE 1 INHIBITIONOF BINDINGOF ID-ANTI-ID BY VARIOUSINHIBITORS Inhibitor@

Specificity

% inhibition

Rabbit No. 1 b

Preimmune Anti-HIV peptide

0” 2

Rabbit No. 2

Preimmune Anti-HIV peptide

0 5

Rabbit No. 3

Preimmune Anti-HIV peptide

0 3

Mouse McAb Ab Mouse McAb B Mouse McAb C

Anti-HIV peptide Anti-HIV peptide Anti-HIV peptide

1 7 4

Chimpanzee

X99

Preimmune Anti-HIV peptide

0 80

Chimpanzee

Xl 83

Preimmune Anti-HIV peptide

0 78

Chimpanzee

Xl 89

Preimmune Anti-SV40 peptide

0 3

735 to 752 gp41 peptide

Nominal antigen of HIV peptide

0

Recombinant

Recombinant

0

gpl60

HIV antigen

a Each inhibitor was used at a final concentration of 10 pg. dThe characteristics of the rabbit anti-gp41 and mouse McAb have been described elsewhere (Chanh eta/., 1986; Dalgleish eta/., 1988). c Represents mean of triplicate determinations.

different species, we examined the ability of polyclonal chimpanzee and rabbit antibodies and mouse monoclonal anti-gp41 peptide antibodies to inhibit the Idanti-Id reaction (Table 1). Neither rabbit anti-gp41 peptide antibodies nor mouse McAb preparations inhibited the Id-anti-Id reaction. At a concentration of 10 pg, chimpanzee X99 and Xl 83 anti-gp41 peptide inhibited Id-anti-Id reactions by 80 and 78O?&respectively. No inhibition was observed with preimmune chimpanzee IgG or chimpanzee anti-SV40 T-ag peptide antibodies. The concentration of X99 and Xl 83 anti-gp41 peptide that inhibited the Id-anti-Id reaction by 50°b was 180 and 2 10 ng, respectively (data not shown). On the basis of these observations and the results above, it was apparent that the anti-Id preparation did not exhibit any internal image characteristics. In viva administration rabbits

of anti-Id into Balbk

mice and

On the basis of previous studies where both noninternal image and internal image anti-Id could trigger an Ag-specific immune response, we decided to examine the in v&o modulating capability of the anti-Id MCI. For

RESPONSE TO HIV

13

this study, we selected a syngeneic system (Balb/c mice) and a xenogeneic system (New Zealand white rabbits) to examine whether this anti-Id could induce an anti-HIV gp41 response specific for peptide 735 to 752 within and across species barriers. Four groups of five mice each were given six i.p. injections of McAb anti-Id as either an alum precipitate alone or a KLH coupled, alum-precipitated preparation. Two groups received MC1 while the other two received a control IgGl anti-Id. A fifth group of mice was immunized with three injections of peptide 735 to 752 coupled to KLH. Two groups of three rabbits each were injected six times with monoclonal anti-Id preparations as an alum precipitate. One group of rabbits received MC1 and the second group received a control monoclonal IgGl antiId specific for HBsAg. Sera from all groups of animals were tested for anti-peptide and anti-HIV activity between 14 and 28 days after the final immunization (Table 2). The group of mice that received MC1 as an alum precipitate alone produced both anti-peptide and antiHIV responses. The mean reciprocal end point antipeptide titers and anti-HIV titers in MCl-alum precipitate-treated mice was fourfold higher when compared with that in mice that received MCl-KLH. Neitherof the two control anti-Id treated groups of mice, nor the two groups of rabbits, showed any detectable anti-peptide or anti-HIV activity. The production of an anti-Id response (Ab-3) in the rabbit antisera to MC1 and control anti-Id was confirmed by employing direct binding analysis following adsorption of the antisera to remove antiisotypic and anti-allotypic specificities (Fig. 2). At a 1: 500 dilution of antisera, the rabbits immunized with MC1 bound MC1 at an OD,,, of 0.31, whereas no reactivity was observed with control rabbit antisera (OD,,o < 0.00) (Fig. 2A). In contrast, the control rabbit antisera bound the control anti-Id at a dilution of 1:500 (Fig. 2B, OD,,,, 0.28) whereas no binding for rabbit anti-MC1 to control McAb anti-Id was observed. These results indicate that the rabbits immunized with MC1 and control anti-Id did produce a specific anti-Id response (Ab-3) to the mouse monoclonal anti-Id preparations. Monoclonal anti-Id induced an Ab-3 response that shares an Id with the chimpanzee Ab-1 A striking serologic difference between the anti-Id and nominal antigen-immunized mice and rabbits was that the anti-Id-treated animals inhibited the Id-anti-Id reactions (ranging from 20 to 40%) whereas no inhibition was obtained in animals immunized with peptide 735 to 752 (nominal antigen) or with control anti-ld-immunized animals (Fig. 3, Table 2). These studies suggest that anti-Id immunization with a noninternal image

14

ZHOU, LOHMAN,

AND KENNEDY

TABLE 2 ANTI-ID INDUCES AN IN Iho ANTI-HIV RESPONSE

Treatment”

Animals

Anti-HIV gp41 peptide (mean f SEM)b

No. of injections

Anti-HIV wl60 (mean + SEM)

% inhibition of Id-anti-Id (range)

Mice

MC1 MC1 -KLH Control anti-Id Control anti-Id-KLH HIV gp41 peptide-KLH

6 6 6 6 3

160-+ 40 40* 10 (20